Ginkgo Biloba: Decades of Research, an Inconsistent Record

Few botanicals have been studied as extensively as Ginkgo Biloba for memory and cognition. That extensive record is exactly why its inconsistent results — including one of the largest trials ever run on a memory supplement — are worth taking seriously rather than skipping past.

Botanical name
Ginkgo biloba
Part used
Leaf extract
Ginkgo Biloba (Ginkgo biloba)

What Ginkgo is

Ginkgo biloba is one of the oldest living tree species, and its leaves have been used in Chinese medicine for centuries. The modern supplement is almost always a standardised leaf extract — most research uses a specific preparation known as EGb 761, standardised to 24% flavone glycosides and 6% terpene lactones — rather than raw leaf material.

This standardisation matters more than for most botanicals: results from studies using EGb 761 do not automatically transfer to a product using an unstandardised or differently processed extract.

The research record — and the GEM trial

Proposed mechanisms include improved microcirculation, antioxidant activity, and inhibition of platelet-activating factor. Decades of smaller trials produced a genuinely mixed picture — some found modest benefits on measures of cognitive function or symptoms of age-related cognitive decline, others found none.

The most important single study is the Ginkgo Evaluation of Memory (GEM) study, a large, NIH-funded, randomised placebo-controlled trial that followed over three thousand older adults for an average of six years to test whether Ginkgo could prevent dementia. It found no significant reduction in the rate of dementia or Alzheimer’s disease compared to placebo. This is one of the largest and longest trials ever run in this category, and its null result carries real weight against smaller, more favourable studies.

How to read the mixed record fairly

Smaller and shorter trials have found modest benefits on specific measures in specific populations. The GEM study specifically tested dementia prevention over years, in generally healthy older adults — a different, harder question than whether Ginkgo modestly affects a cognitive test score over a few months. Both things being true is not a contradiction, but it does mean no claim that Ginkgo prevents cognitive decline is supported by the best available evidence.

Safety — the bleeding-risk caution

The most important caution with Ginkgo

Ginkgo has a well-documented effect on platelet function and is associated with an increased bleeding risk, including case reports of serious bleeding events. Anyone on anticoagulant or antiplatelet medication — including daily low-dose aspirin — should speak to a doctor before taking it. It is standard advice to stop Ginkgo at least two weeks before any scheduled surgery.

  • Seizure disorder. Rare case reports associate high-dose or unprocessed Ginkgo (which can contain a compound called ginkgotoxin) with seizures. Standardised extracts remove most of this compound, but the caution remains standard for anyone with a seizure history.
  • Diabetes medication. Some research suggests a mild effect on blood sugar; monitor if you take glucose-lowering medication.
  • Pregnancy and breastfeeding. Avoid — adequate safety data does not exist.

The GEM study and related research are searchable on PubMed.

Frequently asked questions

Does Ginkgo Biloba prevent dementia?
The largest trial designed to test this specifically — the GEM study, over three thousand older adults followed for an average of six years — found no significant reduction in dementia risk compared to placebo. This is the best current evidence on the question.
Why does the research seem contradictory?
Smaller, shorter trials on specific cognitive measures have found modest benefits; the largest, longest trial designed to test dementia prevention specifically found none. Both results are real; they answer different questions at different scales of evidence.
Is Ginkgo Biloba safe to take?
The most important caution is bleeding risk — Ginkgo affects platelet function, and anyone on blood-thinning medication or with surgery scheduled should speak to a doctor first. It is otherwise generally well tolerated at standard doses.
Does the extract type matter?
Yes. Most research uses a standardised extract (EGb 761, standardised to 24% flavone glycosides and 6% terpene lactones). Results do not automatically apply to unstandardised leaf products.